Palantir Adds Skills to Agents: How Enterprise AI Begins Reusing Analysis Methods

Update Overview

On 2026‑08‑18 Palantir released AIP Analyst 2.0, adding time‑series analysis, Ontology interfaces, export to Foundry resources, and two new reusable‑method features: Skills and Analysis Lookup .

Skills

A Skill is defined as a set of reusable instructions that encode an analysis method. Users create or update a Skill from an active analysis via the Skills UI, and can export the definition as an AIP Skill object. When the Agent receives a new question, it matches the question against stored Skills; if a match is found the Agent loads the Skill and executes the encoded steps on the current data. Because a Skill represents the method rather than a snapshot of results, the same Skill can be invoked repeatedly for similar problems without re‑authoring the workflow.

Analysis Lookup

Analysis Lookup lets the Agent retrieve a historical analysis by its resource identifier (RID) or from a bookmarked list. The retrieved analysis is presented only as a template: the live results from the original run are not reused. When the current task requires the same tools, the Agent re‑executes those tools on up‑to‑date data, ensuring that stale values are not mistaken for the current state.

Method Memory Layer

Together Skills and Analysis Lookup introduce a second layer to Agent Memory. In addition to “knowledge memory” (content retrieved by semantic search), the Agent now retains “method memory” – a record of which resources, Ontology objects, and tool pipelines were used to solve a class of problems, and can repeat that analytical pathway with fresh data.

Ontology Coupling

The mechanisms rely on Palantir’s Ontology system, which unifies data, logic, actions, and security into a single business representation. AIP Analyst queries Object Types, builds Object Sets, runs Ontology SQL, performs aggregations, and renders charts or maps on top of this representation. Because Skills and Analysis Lookup operate on the same Ontology‑based objects and tool calls, they are tightly coupled to the Ontology layer.

Implications

The update does not replace semantic search or retrieval‑augmented generation (RAG); instead it adds a reusable‑method capability that lets enterprise AI accumulate “how to analyze” alongside “what was said”. This extends AI memory from pure content storage to actionable, repeatable analysis patterns.

References

Palantir Foundry Announcements – AIP Analyst now supports AIP Skills, Analysis Lookup, time‑series analysis, Ontology interfaces, and Foundry exports (2026‑08‑18). https://www.palantir.com/docs/foundry/announcements

Palantir Docs – AIP Analyst – Capabilities. https://www.palantir.com/docs/foundry/aip-analyst/capabilities

Palantir Docs – Using AIP Analyst. https://www.palantir.com/docs/foundry/aip-analyst/using-aip-analyst

Palantir Docs – The Ontology system. https://www.palantir.com/docs/foundry/architecture-center/ontology-system